Prepared by the Arid Lands Information Center, Office of Arid Lands Studies, University of Arizona ; Steven L. Hilty, compiler.

The mountains of Lesotho form the catchments for the Lesotho Highlands Water Project (LHWP), which is presently under construction, and their condition will determine the longevity of the LHWP. The mountain rangelands also support an extensive livestock system. However, there is concern that grazing is negatively affecting the mountain vegetation to the detriment of both livestock production and catchment function. Therefore, the impact of environment and grazing on the vegetation was investigated to aid the development of management policy for the conservation of the grazing, floristic and water resources of the mountains. Vegetation surveys were conducted in the mountains in the east (Study Area 1: 2 625 - 3 350 m a.s.l.) and in the west (Study Area 2: 2 240 - 3 125 m a.s.l.). Indirect gradient analysis (IGA) and classification were used to investigate the influence of environment on vegetation pattern. Results of the IGA indicated that variation in species composition in the mountains is related primarily to topographic variation, in particular elevation and aspect. Five vegetation communities were identified in Study Area 1 and seven in Study Area 2. These communities occurred consistently in specific topographic positions in the landscape and were arranged along a temperate/subtropical grass species continuum which was associated with a gradient in elevation and aspect. In Study Area 1, the elevation boundary between the high-lying temperate grasslands and the lower subtropical grasslands corresponded with the generally recognised boundary between the Alpine and Subalpine vegetation belts (viz. c. 2 950 m a.s.l. on northerly aspects and c. 2 750 m a.s.l. on southerly aspects). This boundary was lower in Study Area 2 (viz. c. 2 800 m a.s.l. on northerly aspects and c. 2 300 m a.s.l. on southerly aspects). Vegetation-insolation relationships were investigated in Study Area 1 using a model for simulating solar radiation, temperature and potential evaporation patterns on sloping terrain (RADSLOPE). The spatial distribution of the identified vegetation communities and the ratio of temperate (C₃) and subtropical (C₄) grasses in the sward were related to solar irradiance patterns, as influenced by topography. Results suggest that exposure, which increases with altitude, is probably also an important determinant of vegetation pattern in the mountains. The influence of grazing on the vegetation was studied by examining changes in species composition and cover that were associated with gradients in grazing intensity that exist around cattleposts in the mountains. There was little evidence of a shift in species composition and cover under grazing in the Alpine Belt but there was an identifiable grazing gradient in the Subalpine belt. There, short dense grasslands, dominated by palatable species, degrade to a dwarf karroid shrubland with sparse cover under prolonged, intense grazing. The optimum position along the grazing gradient of the more abundant species was identified. It was proposed that the relative positions, or scores, of these species along the grazing gradient can be used in a weighted scoring procedure to provide an index for monitoring the response of the mountain vegetation to grazing. However, the species’ scores still require verification. The need for monitoring temporal changes in vegetation composition and cover in order to assess the possible effects of the LHWP and other development initiatives was noted. Such monitoring should be undertaken in conjunction with an overall programme to assess the dynamics of the socio-economy in the mountains. Therefore, interdisciplinary monitoring programmes are required to achieve this. These programmes should be focused in a few key study locations rather than spread over a wide area. / Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg.

M.Sc. / The Lesotho Highlands Water Project (LHWP) is bound to adversely affect the ecology of the Orange River system and in particular the highland tributaries. The Senqunyane River (a tributary of the Orange River) would be no exception. Water extraction by means of the proposed Mohale Dam (Phase 1B of LHWP) is expected to alter the downstream habitat of the Senqunyane River, which is alarming as the rare rock catfish (Austroglanis sclateri) occur in these waters. Another concern is the lack of knowledge on the riverine ecology as no research has been conducted in the Senqunyane River downstream from the proposed Mohale Dam wall. In order to gain baseline information on the aquatic habitat, distribution and composition of the fish community as well as habitat parameters of the rock catfish , four surveys (from Spring 1994 to Summer 1996) were conducted in the Senqunyane River, major tributaries and in the Senqu River. This information is needed for future monitoring of downstream alterations to the riverine ecosystem and to determine the habitat preference of A. sclateri. Water quality variables were generally in accordance with the guidelines of Kempster, Hattingh & Van Vliet (1980) and the Department of Water Affairs and Forestry (1993), but excessive levels of un-ionized ammonia and phosphates were recorded at antropogenically disturbed areas. Two distinctly different fish communities were respectively found upstream and downstream from the Semongkoaneng Waterfall as this waterfall serves as a migration barrier. The upstream community consists entirely of Psemlobarbus quathlantbae (Maloti minnow), while the downstream community consists of Barb us amens (Smallmouth yellowfish). Labe() capensis (Orange River mudfish). Austraglanis .cclateri (rock catfish) and Oncorhunchus An ecological investigation of the aquatic vertebrates below the proposed Mohale Darn Wall in the Senqunyane River, Lesotho, with emphasis on the rock catfish (Austroglanis sclateri) al1111111Hry mykiss (Rainbow trout). B. aeneus has the highest-representation within this community while 0. mykiss has the lowest representation. Low numbers of the latter species is probably related to the temperature regime and bottom siltation and therefore the Senqunyane River can be regarded as of a marginal area for the occurrence of trout. A. sclateri also has a low representation within the community which is probably a result of its specific habitat requirements as apposed to 13. aeneus and L. capensis which utilize a wider variety of habitat types. Distinctly different habitat preferences were observed for different A. sclateri lifestages. Recruits preferred backwater pools, juveniles preferred suckles and adults preferred runs. It is clear that juveniles and adults prefer strong current velocities and would probably not do well in conditions of reduced flow. A. sclateri could therefore be regarded as an indicator species in the context of instream flow requirements for the Senqunyane River below the proposed Mohale Dam wall.